agentleFS
GitHubKnown publishergithub/spec-kit/docs/integrations/claude.md

spec-kit / integrations

CLAUDE.md142k starsChanged 5 months agoMITView on GitHub

Install Claude Code before setup. The integration key is claude. This integration is declared multi-install safe. See the shared multi-install rules. For an existing Spec Kit project, use specify integration install claude. See integration management for activation, upgrades, and removal. Skills are installed as .claude/skills/speckit-<command>/SKILL.md. Invoke them in Claude's chat, for example /speckit-specify or /speckit-bug-assess for the installed bug extension. By default every Spec Kit skill installed for Claude Code is model-invocable, so its description is always in Claude's context…

We split this claude.md into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use --integration-options="--events false" to disable Spec Kit event handlers.

    line 65

≈ “Common” is an estimate: at least 1 in 5 of the 25,680 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
claude.md

Claude Code

Install Claude Code before setup. The integration key is claude.

Multi-Install Safety

This integration is declared multi-install safe. See the shared multi-install rules.

Setup

specify init my-project --integration claude

For an existing Spec Kit project, use specify integration install claude. See integration management for activation, upgrades, and removal.

Installed Skills

Skills are installed as .claude/skills/speckit-<command>/SKILL.md. Invoke them in Claude's chat, for example /speckit-specify or /speckit-bug-assess for the installed bug extension.

User-Invocable-Only Skills

By default every Spec Kit skill installed for Claude Code is model-invocable, so its description is always in Claude's context and counts toward the skill-listing budget. In projects with many skills this can push other skills out of the listing. To make the core Spec Kit command skills user-invocable only:

specify init my-project --integration claude --integration-options="--no-model-invocation"

This writes disable-model-invocation: true into the core command skills. Claude can run them only when the user invokes /speckit-<command>, and their descriptions stay out of its context.

  • Extension skills stay model-invocable, because mandatory extension hooks rely on the model invoking them. If an extension's mandatory hook runs a core command (its command: is a core speckit.<command>), leave this option off: the model cannot run that hook while the core skill is user-invocable only.
  • The option is stored with the integration's other options and reapplied by specify integration upgrade claude. Passing --integration-options to upgrade replaces all stored options, so repeat any you want to keep (for example --integration-options="--events false --no-model-invocation"); --integration-options="" clears them. Re-running specify init likewise stores only the options passed to it.
  • Core command skills that a preset overrides, or stops overriding, are reverted to the default by commands that re-register preset skills without reinstalling Claude: commands that install, remove or update presets (including specify bundle commands that carry presets), specify integration use claude, and specify integration switch claude when Claude is installed but not the default. To reapply the option, run specify integration upgrade claude --force while Claude is the default integration (--force is needed because those commands modify installed skill files, and it overwrites local edits to Spec Kit-managed files).

Runtime Events and Workflow Dispatch

The integration supports native events through .claude/settings.json. Use --integration-options="--events false" to disable Spec Kit event handlers; combine it with --no-model-invocation if both settings are wanted. See runtime events.

Workflow steps use claude -p <prompt>, with --model when configured and --output-format json when structured output is requested. This adapter does not add a permission-bypass flag itself.

Download now (ZIP)

Trending

Cooling since its April peak.about the whole repository

Week of Oct 26: 69Week of Nov 2: 72Week of Nov 9: 77Week of Nov 16: 58Week of Nov 23: 52Week of Nov 30: 52Week of Dec 7: 46Week of Dec 14: 46Week of Dec 21: 33Week of Dec 28: 29Week of Jan 4: 33Week of Jan 11: 37Week of Jan 18: 42Week of Jan 25: 41Week of Feb 1: 40Week of Feb 8: 51Week of Feb 15: 53Week of Feb 22: 46Week of Mar 1: 46Week of Mar 8: 48Week of Mar 15: 49Week of Mar 22: 70Week of Mar 29: 94Week of Apr 5: 100Week of Apr 12: 100Week of Apr 19: 80Week of Apr 26: 62Week of May 3: 67Week of May 10: 82Week of May 17: 86Week of May 24: 72Week of May 31: 75Week of Jun 7: 63Week of Jun 14: 62Week of Jun 21: 61Week of Jun 28: 50Week of Jul 5: 47Week of Jul 12: 31Week of Jul 19: 34Week of Jul 26: 30Week of Aug 2: 30Week of Aug 9: 29Week of Aug 16: 34Week of Aug 23: 30Week of Aug 30: 34Week of Sep 6: 34Week of Sep 13: 32Week of Sep 20: 35Week of Sep 27: 31Week of Oct 4: 23Week of Oct 11: 22Week of Oct 18: 21Week of Oct 25: 21Week of Nov 1: 17
Oct 2025May 2026Nov 2026
Weekly Google searches for “spec kit”, the term we track for the whole github/spec-kit repository, not this file alone. 100 is its busiest week.

Who's talking about github/spec-kit

The biggest 5 of 113 mentions linking to the repository that our crawl found. Each line is their own headline.

See all 113

Talked about on Bluesky, Hacker News, YouTube, GitHub and the web (113 times).

On Bluesky · 50

45 more on Bluesky

On Hacker News · 18

13 more on Hacker News

On YouTube · 17

12 more on YouTube

On GitHub · 15

10 more on GitHub

On the web · 13

8 more on the web

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.